Urban land cover classification using worldview-2 images and C4.5 algorithm

Mapping of urban land cover using remote sensing technology has been widely explored, especially with the recent availability of high resolution images and object-based analysis techniques. This study uses the InterIMAGE software and WorldView-2 sensor imagery, two recent technologies useful for urban studies, to classify land cover in a metropolitan area of São Paulo, Brazil. Therefore, this work aims to compare the classification performance of two urban land cover thematic maps produced by the object-based image analysis (OBIA) and C4.5 data mining algorithm. The results showed that data mining technique presented classification performance similar to the OBIA method but in a reduced computational time once the entire processing is optimized.